Human Correspondence Consensus for 3D Object Semantic Understanding
arXiv:1912.12577
Abstract
Semantic understanding of 3D objects is crucial in many applications such as object manipulation. However, it is hard to give a universal definition of point-level semantics that everyone would agree on. We observe that people have a consensus on semantic correspondences between two areas from different objects, but are less certain about the exact semantic meaning of each area. Therefore, we argue that by providing human labeled correspondences between different objects from the same category instead of explicit semantic labels, one can recover rich semantic information of an object. In this paper, we introduce a new dataset named CorresPondenceNet. Based on this dataset, we are able to learn dense semantic embeddings with a novel geodesic consistency loss. Accordingly, several state-of-the-art networks are evaluated on this correspondence benchmark. We further show that CorresPondenceNet could not only boost fine-grained understanding of heterogeneous objects but also cross-object registration and partial object matching.
arXiv comment18 pages; ECCV 2020
References in corpus (9)
- Adam: A Method for Stochastic Optimization
- End to End Learning for Self-Driving Cars
- Dense Object Nets: Learning Dense Visual Object Descriptors By and For Robotic Manipulation
- Discovery of Latent 3D Keypoints via End-to-end Geometric Reasoning
- PointConv: Deep Convolutional Networks on 3D Point Clouds
- Sim-to-Real Reinforcement Learning for Deformable Object Manipulation
- SPair-71k: A Large-scale Benchmark for Semantic Correspondence
- Visual Attribute Transfer through Deep Image Analogy
- Self-supervised Learning of Dense Shape Correspondence